CSCI 335 - Artificial Intelligence

Spring 2020

Programming Project #5: Handwriting Recognition with Self-Organizing Maps

Overview

You will implement the self-organizing map. You will experiment with different configurations of this type of neural network to find the best results for a handwriting recognition problem.

Setup

Download som.zip.
Unzip the files into the IntelliJ Java project you employed for Project #4.

Programming Assignment

The new files are in the handwriting.learners.som package. The file SOM6.java in the handwriting.learners package is an example of a fully configured classifier. Create your own with different sized maps.

handwriting.learners.som
- SelfOrgMapTest.java: Contains jUnit tests for all of the above methods you must complete. For each method to complete, its unit test from this file is stated below:
- FloatDrawing.java: Represents a SOM node. Complete the following methods:
  - euclideanDistance (passes testEuclideanDistance())
  - averageIn() (passes testAvg())
- SOMPoint.java: Used for referencing an output node of the SOM. No code to complete, but it is used heavily in SelfOrgMap.java.
- SelfOrgMap.java: Skeletal implementation. Complete the following methods:
  - computeDistanceWeight() (passes testDistanceWeight())
  - effectiveLearningRate() (passes testLearningRate())
  - bestFor() and train() (passes testTrain())
- SOMRecognizer.java: Transforms an unsupervised SOM into a classifier. Complete the following method:
  - findLabelFor() (passes testLabel())

For your experiments, use the training and testing files you created (or borrowed) as part of Project 4.

Visualization

The nature of the self-organizing map lends itself to very lucid visualization. An example of a 6x6 SOM trained on the letters A and H is given below.

Presentations

Thursday, March 5: Short presentation detailing progress thus far. Special emphasis on describing your representation scheme and distance function.
Tuesday, March 10: Presentation summarizing results from the paper.

Paper

When you are finished with your experiments, write a paper summarizing your findings. Include the following:

An analysis and discussion of your data. (Be sure to include the data as well.)
An analysis and discussion of your visualizations.
How well does the self-organizing map perform for this task?
What effect did variations in the map size have on the results?
How do the results compare with multi-layer perceptrons?
Beyond the actual results, what other issues are noteworthy?

Grading criteria

Achievement	Points
Passes `testEuclideanDistance()`	3
Passes `testAvg()`	3
Passes `testDistanceWeight()`	3
Passes `testLearningRate()`	3
Passes `testTrain()`	8
Passes `testLabel()`	5
Ran fourteen experiments, from 2 to 8 letters, alternating training/testing sets	20
Paper quantitatively assesses each of the fourteen experiments	4
Paper includes at least one visualization, and insightfully discusses its implications	5
Paper insightfully discusses how the SOM performs with respect to their data set	4
Paper insightfully discusses how the SOM performs in comparison with the multi-layer perceptron	4
Paper insightfully discusses the distinctions in performance as the number of letters increases	4
Paper insightfully discusses the impact of variations in the map size	4